Introduction
Fatty acids (FAs) are essential nutrients for life, but they are double-edged swords in the battle for good health. While polyunsaturated fatty acids (PUFA) can offer significant health benefits like reducing inflammation and supporting heart health, saturated fatty acids (SFAs) can increase the risk of cardiovascular disease (CVD) and inflammation when consumed in excess. SFAs are well-known components of most animal fat and processed foods, such as those deep-fried in hydrogenated oil and sausage. Because of the adverse effect of FAs on health [1,2], guidelines have advocated reducing the intake of SFAs to improve overall health and especially the risk of CVD [3–5]. However, the beliefs for a diet restrictive of SFAs have been challenged in several recent studies [6,7]. In addition, the results from a recent meta-analysis using prospective cohort studies demonstrated an insignificant relationship between SFAs and CVD or coronary artery disease [8–10]. These inconsistent results could be related to the heterogeneity of the participants, the complexity of food ingredients, and diverse individual dietary habits.
Based on the number of double bonds, unsaturated FAs are discriminated from SFAs, and the effect as a nutrient was differentiated [3]. After the periods for emphasizing the restriction of dietary SFAs, long-chain omega (n)-3 PUFAs such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have been regarded as significant nutrients for promoting cardiovascular health since the early 1970s [11]. Despite the favorable reports of PUFAs as beneficial FAs, especially for cardiovascular health, it is unclear whether these results can be applied to all populations regardless of kidney function. Nevertheless, we expect that dietary intake of PUFAs could be a helpful guide based on decreased inflammation and oxidative stress in patients with kidney dysfunction [12].
Decreased kidney function is one of the leading factors in poor clinical outcomes. The presence of kidney disease significantly increases the global health burden, including CVD and mortality [13]. In addition, considering the altered pathophysiology in the metabolism of nutrients, this population requires a different approach regarding dietary counseling for nutrients compared to the general population with normal kidney function [14]. Although several randomized clinical trials using a PUFA supplement showed no significant difference in the reduction of major cardiovascular outcomes in high-risk patients compared to the placebo group, dietary intake of N3 PUFA has been associated with a reduced risk of myocardial infarction in patients with end-stage kidney disease [15–18]. However, there is a lack of evidence for the clinical impact of each FA subtype on all-cause mortality according to kidney dysfunction status. In this regard, we aimed to investigate the different impacts of dietary FAs on all-cause mortality according to kidney function.
Methods
Study populations and data acquisition
Participants aged ≥18 years old who registered for the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2016 were eligible for the study. We excluded participants with an implausible range of total energy intake (<500 or ≥8,000 kcal/day for males, <500 or ≥5,000 kcal/day for females) and no data for kidney function. We obtained the following data: anthropometric data (height, weight, and blood pressure), laboratory data (blood urea nitrogen [BUN], serum glucose, uric acid, total cholesterol, estimated glomerular filtration rates [eGFRs], urine albumin-to-creatinine ratio [UACR]), and dietary data (total energy, carbohydrates, protein, fat, and each specific FA).
Dietary assessment
FA and other nutrient intakes were estimated using a 1-day 24-hour dietary recall. The participants reported the types and amounts of foods and beverages consumed in the 24 hours before the interview by trained interviewers. Among the nutrient datasets, we used total energy, carbohydrate, protein, and fat intakes. SFAs were divided into eight subtypes: butanoic acid (C 4:0), hexanoic acid (C 6:0), octanoic acid (C 8:0), decanoic acid (C 10:0), dodecanoic acid (C 12:0), tetradecanoic acid (C 14:0), hexadecanoic acid (C 16:0), and octadecanoic acid (C 18:0). PUFAs were categorized into seven subtypes: octadecadienoid acid (n-6, C 18:2), octadecatrienoic acid (n-3, C 18:3), octadecatetraenoic acid (n-3, C 18:4), eicosatetraenoic acid (n-6, C 20:4), EPA (n-3, C 20:5), docosapentaenoic acid (DPA) (n-3, C 22:5), and DHA (n-3, C 22:6). We also calculated four additional FAs: total SFA, total PUFA, omega-6 FAs, and omega-3 FAs. All macronutrients were calculated as percentages of total energy intake (% of energy) by dividing energy intake from each macronutrient by daily total energy intake (kcal).
To investigate the association between dietary FAs and all-cause mortality independent of other dietary factors, the Healthy Eating Index (HEI)-2020 score was calculated which was designed to evaluate adherence to the 2020–2025 Dietary Guidelines for Americans [19]. The HEI-2020 is composed of nine adequacy components including total fruits, whole fruits, total vegetables, greens and beans, whole grains, dairy, total protein foods, seafood and plant proteins, and FAs (PUFA + monounsaturated FAs) and four moderation components including refined grains, sodium, added sugars, and SFA. The original HEI-2020 score ranged from 0 to 100, but we excluded the components related to FAs from the total score and the final score ranged from 0 to 80. In addition, consumption of FA supplements (yes/no) such as omega-3 FAs and fish oil was evaluated.
Outcome assessment and subgroup analysis
As a primary clinical outcome, we assessed all-cause mortality. Mortality data was obtained from the National Center of Health Statistics, which was linked to the National Death Index up to December 31, 2016. Survival time was calculated from the date of the clinical examination to the date of death.
We performed subgroup analysis using eGFR, and the group was divided into eGFR ≥90 mL/min/1.73 m2, 60–90 mL/min/1.73 m2, and <60 mL/min/1.73 m2. An eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.
Statistical analysis
We demonstrated a mean ± standard deviation for continuous variables and a number with a percentage for categorical variables. We performed the Student t tests and chi-square tests to compare the two groups. Two-sided p-values were derived by setting the significance level at 0.05. All FAs except octadecatetraenoic acid (C 18:4) were divided into quartile ranges, and the first quartile was regarded as a reference value. The intake of octadecatetraenoic acid (C 18:4) was close to 0%, and it was divided into tertiles. The first tertile was regarded as a reference value. We performed Cox regression analysis to evaluate the association of dietary FAs with mortality. All variables were found to satisfy the proportional hazard assumption. We used age, sex, ethnicity, education status, household income status, body mass index (BMI), alcohol consumption, smoking, systolic blood pressure, serum albumin, serum glucose, total cholesterol, serum uric acid, UACR, total energy intake, consumption of FA supplements, and the HEI-2020 score excluding FA components as covariates in the multivariable analysis. A restricted cubic spline model was used to illustrate the association between dietary FA intake and the risk of all-cause mortality. All statistical analyses were performed using IBM SPSS version 20.0 (IBM Corp.) and SAS version 9.4 (SAS Institute).
Results
Study populations
After excluding ineligible participants, 44,332 individuals from the total 101,316 potential participants were finally included in the study (Supplementary Fig. 1, available online). There were 26,246 (59.2%), 13,937 (31.4%), and 4,149 participants (9.4%) with eGFR ≥90, 60–90, and <60 mL/min/1.73 m2, respectively. In comparing each group according to the eGFR category, the lowest eGFR group members were older, and a higher proportion were ex-smokers, non-alcohol drinkers, who had a lower education status, greater BMI status (p < 0.05 for all). Additionally, they had more members of white and black races with higher systolic blood pressure; higher serum levels of BUN, glucose, uric acid, and UACR; and lower serum levels of albumin and eGFR (p < 0.001 for all) (Table 1).
Dietary intake according to the kidney function
Overall total energy intake was 2,129.93 ± 964.16 kcal, and it was significantly different according to the groups (2,247.72 ± 1,001.86 in the eGFR ≥90 mL/min/1.73 m2 vs. 2,035.37 ± 904.68 in the eGFR 60–90 mL/min/1.73 m2 vs. 1,702.51 ± 730.71 in the eGFR <60 mL/min/1.73 m2). The amount of total energy intake, and macronutrient intakes such as carbohydrates, protein, and fat as absolute intake (g/day) was incrementally decreased in participants with the lower eGFR group (Supplementary Table 1, available online). It was observed that the lower the kidney function, the higher the HEI. Notably, the group with an eGFR 60–90 mL/min/1.73 m2 had the highest proportion of individuals with a history of supplemental intake (p < 0.001 for all) (Table 1). In terms of the proportion of total SFA (% of energy) was significantly increased according to the decreasing eGFR groups (10.69% of energy in eGFR ≥90 mL/min/1.73 m2, 10.82% of energy in eGFR 60–90 mL/min/1.73 m2, and 10.93% of energy in eGFR <60 mL/min/1.73 m2), and the proportion of total PUFA was also slightly increased according to the decreasing eGFR groups (7.24% of energy in eGFR ≥90 mL/min/1.73 m2, 7.57% of energy in eGFR 60–90 mL/min/1.73 m2, and 7.66% of energy in eGFR <60 mL/min/1.73 m2) (Supplementary Table 1, available online). Hexadecanoic acid (C 16:0) and octadecanoic acid (C 18:0) were the main components among the dietary SFAs, and the proportion of both SFAs was higher in participants with lower eGFRs. Octadecadienoic acid (C 18:2) and octadecatrienoic acid (C 18:3) were the main components of dietary PUFAs, and they were higher in the lower eGFR group. The dietary proportion of omega-6 FAs was more prevalent than omega-3 FAs, and both FAs were higher in participants with lower eGFR (Supplementary Table 2, available online).
Impact of dietary fatty acids on all-cause mortality
There were 6,934 deaths (15.7%) during 129.0 ± 62.4 months of follow-up. The most common cause of death was cardio-cerebrovascular disease (n = 2,189, 5.0%), followed by malignant disease (n = 1,509, 3.4%) (Supplementary Table 3, available online).
Higher intake of total SFAs incrementally increased the risk of all-cause mortality based on the reference of 10.0% of energy, the mean intake of SFA (Supplementary Fig. 2A, available online). After adjustment with such variables, the highest quartile of total SFAs (adjusted hazard ratio [aHR], 1.10; 95% confidence interval [CI], 1.01–1.20) showed increased risk compared to the lowest quartile of intake. In addition, among each SFA, the highest quartile of tetradecanoic acid (C 14:0) (aHR, 1.09; 95% CI, 1.01–1.19 in the 4th quartile) showed increased risk compared to the lowest quartile of intake (Table 2). Meanwhile, a higher intake of total PUFAs incrementally decreased the risk of all-cause mortality based on the reference 7.0% of energy, the mean intake of PUFAs (Supplementary Fig. 2B, available online). In addition, most PUFAs, except eicosatetraenoic acid (C 20:4) and DPA (C 22:5), showed a significantly decreased risk in the higher quartile of intake. Additionally, a higher intake of omega-6 and omega-3 FAs showed a significantly reduced risk of all-cause mortality (Table 3).
Impact of dietary fatty acids on all-cause mortality according to kidney function
The impact of dietary SFA on all-cause mortality was different according to the three groups divided by eGFR categories. The negative effect of SFAs was more prominent in participants with eGFR 60–90 mL/min/1.73 m2 (Fig. 1B) than in other groups (Fig. 1A, C). For participants with eGFR ≥90 mL/min/1.73 m2 or <60 mL/min/1.73 m2, there was no hazard effect of all types of SFAs, even total SFA (Fig. 2A, C). However, a higher intake of hexadecanoic acid (C 16:0) and octadecanoic acid (C 18:0) increased the risk of all-cause mortality in participants with eGFR 60–90 mL/min/1.73 m2 (Fig. 2B). The highest quartile of total SFA intake also significantly increased the risk of all-cause mortality (aHR, 1.20; 95% CI, 1.06–1.36) in participants with eGFR 60–90 mL/min/1.73 m2.
Similarly, the decreased risk for all-cause mortality with a higher intake of PUFAs was more prevalent among participants with eGFR 60–90 mL/min/1.73 m2 (Fig. 3B) than other groups (Fig. 3A, C). Among the PUFAs, the highest quartile group of octadecatrienoic acid (C 18:3) (aHR, 0.83; 95% CI, 0.69–0.99) showed a significantly decreased risk of all-cause mortality in participants with eGFR ≥90 mL/min/1.73 m2 (Fig. 4A). Except for octadecatrienoic acid (C 18:3), there was no significant association between all-cause mortality and higher intake of total PUFAs, omega-6 FAs, and omega-3 FAs in participants with eGFRs ≥90 mL/min/1.73 m2 (Fig. 4A). In addition, there were no significant PUFAs reducing the risk of all-cause mortality according to the dietary intake in participants with eGFR <60 mL/min/1.73 m2 (Fig. 4C). Meanwhile, several PUFAs such as octadecadienoic acid (n-6, C 18:2), and octadecatrienoic acid (n-6, C 18:3) showed that higher intake significantly decreased the risk of all-cause mortality in participants with eGFR 60–90 mL/min/1.73 m2 (Fig. 4B). In addition, the highest quartile of total PUFAs (aHR, 0.85; 95% CI, 0.75–0.97), omega-6 FAs (aHR, 0.85; 95% CI, 0.75–0.96), and omega-3 FAs (aHR, 0.85; 95% CI, 0.75–0.96) significantly decreased the risk of all-cause mortality.
Impact of dietary fatty acids on cardio-cerebrovascular mortality
There were 2,189 deaths (5.0%) due to cardio-cerebrovascular disease during a follow-up period of 129.0 ± 62.4 months. According to the kidney function groups, there were 387 (1.5%), 987 (7.1%), and 815 deaths (19.6%) in eGFR >90, 60–90, and <60 mL/min/1.73 m2, respectively.
Although individual SFAs did not significantly affect cardio-cerebrovascular mortality, the highest quartile of total SFA intake increased the risk by 1.2 times compared to the 1st quartile (Supplementary Table 4, available online). In the case of PUFAs, the highest quartile intake of octadecadienoic acid (C 18:2) (aHR, 0.85; 95% CI, 0.73–0.99) and EPA (n-3, C 20:5) (aHR, 0.81; 95% CI, 0.70–0.93) significantly reduced the risk of cardio-cerebrovascular mortality compared to the 1st quartile. Similar results were also observed for omega-6 and total PUFA intake (Supplementary Table 5, available online).
In the subgroup analysis according to kidney function, the harmful effect of SFA was observed only in the eGFR 60–90 mL/min/1.73 m2 group, where the highest quartile of octadecanoic acid (C 18:0) (aHR, 1.27; 95% CI, 1.01–1.60) and total SFA (aHR, 1.28; 95% CI, 1.03–1.60) intake significantly increased the risk of cardio-cerebrovascular mortality (Supplementary Fig. 3, available online). Interestingly, contrary to the results for all-cause mortality, the beneficial effect of PUFA was particularly pronounced in the eGFR <60 mL/min/1.73 m2 group (Supplementary Fig. 4, available online). Specifically, for octadecadienoic acid (n-6, 18:2) and omega-6 FAs, the 3rd and 4th quartiles significantly reduced the risk of cardio-cerebrovascular mortality compared to the 1st quartile. For total PUFA, the 4th quartile significantly reduced the risk compared to the 1st quartile (aHR, 0.75; 95% CI, 0.58–0.97) (Supplementary Fig. 4C, available online).
Discussion
FAs are essential nutrients, and many supplements have been developed and advertised worldwide. Considering the lack of data to evaluate the impact of each component of dietary FAs on all-cause mortality, we tried a new approach to identify these effects by kidney function. Dietary FAs showed different associations with all-cause mortality according to baseline kidney function. Most SFAs and PUFAs showed an insignificant association between the intake amounts and all-cause mortality in participants with eGFR ≥90 or <60 mL/min/1.73 m2. However, the harmful effects of SFAs and the beneficial effects of PUFAs were prevalent among participants with eGFR 60–90 mL/min/1.73 m2.
Deteriorated kidney function influences changes in FA metabolism and is related to mitochondrial dysfunction and cellular damage [20]. These changes in FAs might not only have a negative effect on the heart but also contribute to the progression of kidney damage. In contrast to the significance of FAs as an essential nutrient, there is a lack of data for evaluating these impacts on participants with lower eGFR. Following the guidelines, encouraging the intake of PUFAs, especially omega-3 FAs, is recommended only for the lipid profile, not for CVD or mortality for patients with stage 5 chronic kidney disease (CKD) [21]. Moreover, these recommendations were focused on advanced CKD patients, not on overall patients with kidney dysfunction.
In addition, omega-3 FAs have been found to be a beneficial nutrient to reduce the risk of major chronic diseases such as diabetes and CVD [22,23]. Moreover, several observational studies reported that a higher intake of omega-3 FAs is related to a lower risk of all-cause mortality [24–26]. However, the association between dietary intake of omega-3 FAs and all-cause mortality was different according to age, sex, and ethnicity [27,28]. For the whole population, a higher intake of omega-3 FAs incrementally decreased the risk of all-cause mortality in this study. However, this association was attenuated in participants with eGFR ≥90 or <60 mL/min/1.73 m2, and there was a significant association in participants with eGFR 60–90 mL/min/1.73 m2 only. Our results indicate that dietary intake affects all-cause mortality differently depending on the baseline eGFR, with a substantial effect seen in those with marginal kidney function (eGFR 60–90 mL/min/1.73 m2).
Dietary omega-6 FAs make up a major portion of PUFAs, and most Americans eat on average approximately 10 times more omega-6 FAs than omega-3 FAs. Octadecanoic acid (linoleic acid, LA) and eicosatetraenoic acid (arachidonic acid, AA) are the main sources of omega-6 FAs. LA is synthesized in plants, and it plays a role in reducing total blood cholesterol and low-density lipoprotein cholesterol, resulting in improved cardiovascular outcomes [29–31]. Another main component of omega-6 FAs, AA, is commonly found in foods of animal origin, such as meats, eggs, and offal. AA also plays an important role in human health, especially in brain development and cognitive function [32,33]. Despite these advantages, there was a lack of evidence to support that increasing omega-6 FAs is associated with a reduced risk of death [34]. We found that a higher intake of omega-6 FAs significantly decreased the risk of all-cause mortality in the general population in this study. In addition, it was more prevalent in participants with marginal kidney function. In this regard, we suggested that a beneficial effect of a higher intake of omega-6 FAs was expected only in participants with an eGFR 60–90 mL/min/1.73 m2.
In contrast to the beneficial effect of PUFAs, the major component of SFA, hexadecanoic acid (C 16:0), showed trends in increased risk for all-cause mortality. Hexadecanoic acid is found naturally in palm oil, butter, cheese, and meat, and it has harmful effects on lipid profiles and atherosclerosis risk [35,36]. Longer chain octadecanoic acids (C 18:0), the second most abundant SFA, are rich in cocoa butter, mutton tallow, and beef tallow. Unlike other SFAs, octadecanoic acid appears to have some beneficial effects on health, with reduced blood pressure and improved heart function [37,38]. Nevertheless, we found that a higher intake of octadecanoic acids (C 18:0) was significantly associated with an increased risk for all-cause mortality. Based on these multifactorial associations, the association between SFA and all-cause mortality has been inconsistently represented [8,39]. In this study, we found that the impact of higher intake of these two major SFAs was different according to kidney function. The negative effect of higher intake of SFA was prevalent in participants with marginal kidney function with eGFR 60–90 mL/min/1.73 m2 only.
The insignificant association between SFA and mortality in participants with eGFR ≥90 mL/min/1.73 m2 may primarily be due to the low number of outcomes in this group. Specifically, the outcome rates were 1,623 (6.2%), 3109 (22.3%), and 2,202 (53.1%) in the eGFR ≥90, 60–90, and <60 mL/min/1.73 m2, respectively. Furthermore, considering the eGFR distribution within the eGFR ≥90 mL/min/1.73 m2 group, with a median of 108.7 and an interquartile range of 99.1–121.1 mL/min/1.73 m2, it is noteworthy that a quarter of the patients had an eGFR exceeding 120 mL/min/1.73 m2, indicating a pattern that may warrant consideration of hyperfiltration, which could be related to malnutrition. Decreased muscle mass with malnutritional status is closely related to glomerular hyperfiltration. In this case, a higher intake of SFA could reduce the harmful effect of SFA based on the supplementation of nutrition. To explore this association, we used BMI as an adjustment variable, and the result was unchanged. Nevertheless, to consider the impact of hyperfiltration, further evaluation for a more detailed divided subgroup would be necessary. Meanwhile, the substantial effect of kidney dysfunction on all-cause mortality could attenuate the impact of dietary FAs on all-cause mortality in participants with eGFR <60 mL/min/1.73 m2. Contrary to these insignificant results in participants with eGFR ≥90 or <60 mL/min/1.73 m2, the dietary impact was prominent in participants with eGFR 60–90 mL/min/1.73 m2. This finding is particularly significant because it indicates which population would benefit most from dietary intervention.
In the physiological pathway, dietary FAs are subdivided into three broad classes according to the degree of unsaturation. In addition, FAs are also differentiated by the configuration of the double bond and the length of the chain. They are easily influenced by food processing, particularly during partial hydrogenation. Moreover, each component of FAs closely interacts with and shares a common pathway [40]. Therefore, it is difficult to evaluate the independent effect of each component on health. Nevertheless, based on an epidemiological study using a large nationwide cohort, we suggest overall evidence of the beneficial or harmful effects of dietary patterns.
This study represented independent associations between dietary FAs and all-cause mortality according to kidney function. Our findings will be helpful in dietary counseling with regard to FAs, especially for participants with marginal kidney dysfunction. However, this study has several limitations. This is an observational study, and self-reported dietary assessment is prone to measurement errors such as recall bias. We did not consider FA intake from dietary supplements because this study tried to focus on FA intake through the ordinary diet. We classified and analyzed the association between dietary FAs and all-cause mortality according to kidney function, but we could not consider the various underlying diseases or concomitant drug use. To clarify the effects of dietary SFAs and PUFAs in participants with kidney dysfunction, a more detailed randomized clinical trial needs to be performed.
The beneficial effect of a higher intake of PUFAs and the negative effect of a higher intake of SFAs were prominent in participants with eGFR 60–90 mL/min/1.73 m2. In this regard, our findings suggest dietary counseling of FAs should be warranted in these populations with potential risk for progression to advanced kidney disease.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement table 1
Print
